Prostate cancer is the most commonly diagnosed cancer in American men and will kill approximately 27,050 men in 2007. Androgen ablative therapy and docetaxel prolong survival and improve patients'quality of life but metastatic prostate cancer continues to cause significant suffering and death and novel treatments are required. Deregulation of the PTEN/PI3K/AKT pathway is common in advanced prostate cancer and an attractive target for therapy. Inhibition of mTOR, an effector arm of AKT signaling, is sufficient to reverse neoplastic prostate phenotypes in transgenic mice due to prostate-targeted loss of PTEN or AKT activation. However, emerging data from clinical trials suggest that while mTOR inhibition results in stable disease for 40% of men, objective responses are uncommon and the impact on survival in an unselected population of men with advanced prostate cancer remains unclear. This proposal will determine if our established methods of immunohistochemical and expression analysis can identify tumors with mTOR activity, predict stable disease in individual patients treated with mTOR inhibition, and test for associations between alternative pathway or chemotherapy response signatures and resistance to mTOR inhibition. In Aim 1, expression "signatures" of mTOR activity will be developed across a range of prostate cancer models and human tumors to determine how cellular and genetic context impact the ability of mTOR signatures to correlate with mTOR activity. In Aim 2, the immunohistochemical measures and the best mTOR response signatures correlating with mTOR activity from in Aim 1 will be tested for their accuracy in predicting response to mTOR inhibition on prostate cancer samples from a phase II trial during which patients will be treated with RAD001 (an mTOR inhibitor). In Aim 3, by performing immunohistochemical and expression analysis on CT-guided biopsies of metastatic disease prior to and following RAD001 we will determine how changes in measures of mTOR activity and alternative pathways and/or predictive signatures correlate with clinical resistance. Candidate pathways will be tested for their impact on mTOR inhibition using genetically- defined, prostate epithelial cells transformed with activation of the PTEN/PI3K/AKT pathway. As for many targeted therapies, the clinical results of single agent mTOR inhibition in an unselected population of men with metastatic prostate cancers appear modest despite the established biological importance of this pathway in the disease. The overall goal of the work herein proposed is to improve the care of men with prostate cancer by providing predictive markers and identify potentially synergistic combination therapy. While focused on mTOR response in metastatic prostate cancer, the methods developed and tested in this proposal can be broadly applied.